Cagrilintide: Fourteen Trials, And A 0.16-Point Benefit

A Different Hormone Entirely

Everything else in this area works on GLP-1, or GLP-1 plus something adjacent. Cagrilintide does not. It is an amylin receptor agonist.

Amylin is co-secreted with insulin and acts on satiety and gastric emptying through a different pathway. That is the rationale for combining it with semaglutide rather than replacing it: two routes to the same effect, so the combination might do more than either alone. The combination product is called CagriSema.

The logic is sound and the compound is a serious pharmaceutical programme. Which is what makes the trial results worth reading closely, rather than dismissing the way the no-evidence peptides can be.

The Trial That Actually Answers The Question

Most trials compare a drug against placebo, which tells you whether it does anything. REIMAGINE 2 did something more useful: it compared the combination against semaglutide on its own.

That is the question that matters. Semaglutide already exists. The only reason to add a second compound is if the pair beats the one.

The design is strong: randomised, double-blind, placebo- and active-controlled, 30 countries, 68 weeks, 2,713 participants, of whom 95.7% completed. Six arms. Funded by Novo Nordisk, which makes both components.

Primary endpoint, change in HbA1c at week 68:

• CagriSema (2.4 mg each): −1.91 percentage points
• Semaglutide 2.4 mg alone: −1.75 percentage points

Estimated treatment difference: −0.16 percentage points (95% CI −0.27 to −0.05), p=0.0035.

Sit With That Number For A Moment

0.16 percentage points.

That is the entire added benefit of cagrilintide, measured in the trial designed by the manufacturer to demonstrate it, at the top dose, over 68 weeks, in 2,713 people.

It is statistically significant, and the confidence interval excludes zero, so the effect is real rather than noise. The paper's interpretation — that the combination was superior and the findings support its added benefit — is accurate.

“Superior” is a statistical word. It means the difference is unlikely to be chance. It carries no information at all about whether the difference is large enough to matter to a person, and with 2,713 participants a trial can detect differences far smaller than anyone would notice.

For scale: an HbA1c of 8.2% falling to 6.45% versus falling to 6.29%. Both are a substantial improvement. The gap between them is the part cagrilintide contributed.

This is the same distinction as the SELECT trial's 20% versus 1.5 points, arriving from the opposite direction: there a real effect was made to sound enormous, here a very small one is described with a word that sounds decisive.

What The 0.16 Points Cost

The same table carries the other side. Adverse events were reported in:

• 86.9% on CagriSema (2.4 mg each)
• 81.2% on semaglutide 2.4 mg alone
• 70.5% on placebo

So adding cagrilintide bought 0.16 percentage points of HbA1c and cost roughly 6 percentage points more participants reporting an adverse event. Gastrointestinal disorders were the most common in every active arm, as with the rest of this class.

Whether that trade is worth making is a clinical judgement about a specific person, and there are people for whom every fraction of a point of HbA1c matters. It is not a judgement anyone can make from a product page.

And note what this comparison is not. It is a glycaemic control trial in type 2 diabetes. It is not a weight-loss trial, which is what cagrilintide is mostly sold on.

Nobody Is Really Testing It On Its Own

Here is the detail that matters most to anyone thinking of buying cagrilintide by itself, and it is visible only in the arm sizes.

REIMAGINE 2 randomised:

• 603 to CagriSema 2.4 mg each
• 605 to semaglutide 2.4 mg
• 595 to CagriSema 1.0 mg each
• 609 to semaglutide 1.0 mg
• 152 to cagrilintide 2.4 mg alone

The standalone arm is a quarter the size of the others. That is not an oversight; it is what a supporting arm looks like. The programme is built to test a combination, and cagrilintide alone is present to interpret the combination rather than to be evaluated on its own merits.

So the fourteen randomised trials that make cagrilintide the best-evidenced peptide in this category are, overwhelmingly, trials of cagrilintide plus semaglutide. Someone buying cagrilintide on its own is citing an evidence base built around a product they are not taking.

Cagrilintide also has no standalone approval. The combination is the commercial product, and it is a prescription medicine.

What This Says About The Whole Category

Put the three together and the shape of this field is visible:

• BPC-157: 226 papers, zero randomised trials, and a category 2 listing from the FDA.
• Retatrutide: excellent randomised evidence, −24.2% at 48 weeks, and not lawfully purchasable anywhere.
• Cagrilintide: the most randomised evidence of any peptide sold online — and the trial designed to show its benefit shows 0.16 percentage points, in combination, in diabetes.

There is no version of this where a reader wins by buying the peptide. Either there is no evidence, or the evidence belongs to a product they cannot get, or the evidence is real and much smaller than advertised.

That is not an argument that peptide research is worthless. Cagrilintide and retatrutide are serious programmes and one or both may become genuinely useful approved medicines. It is an argument that the gap between a development programme and a vial on a website is the entire thing, and reading the trial does not close it.

Meanwhile, for anyone already on an approved drug in this class, the variable that actually changes what you keep is unchanged and unglamorous: protein and resistance training.